You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
It might be a good idea to use or emulate the HessenbergQ type from LinearAlgebra: this is a type to encode all of the Householder reflectors, and provides specialized methods to multiply vectors and matrices by Q or Q', as well as a Matrix conversion to get the ordinary matrix (which would replace your getQ function).
Ideally, you could just use the HessenbergQ object as-is, as long as you encode your householder reflectors in the same way. (This also allows you to use LAPACK routines to multiply by the reflectors.)
The text was updated successfully, but these errors were encountered:
I adapted the Hessenberg type so that it admits a Tridiagonal matrix as reduced matrix. The type returned by hessenberg(A) is now a standard Hessenberg type.
It might be a good idea to use or emulate the
HessenbergQ
type from LinearAlgebra: this is a type to encode all of the Householder reflectors, and provides specialized methods to multiply vectors and matrices byQ
orQ'
, as well as aMatrix
conversion to get the ordinary matrix (which would replace yourgetQ
function).Ideally, you could just use the
HessenbergQ
object as-is, as long as you encode your householder reflectors in the same way. (This also allows you to use LAPACK routines to multiply by the reflectors.)The text was updated successfully, but these errors were encountered: